[RPRWG] MR's in technical review state
Here is my input on MR's 0025, 0034, and 0037.
Gary
__________________________
MR0025: 7.5.7.2.4 - Table 7.11 and Table 7.12 make use of an undefined constant, NORM
Proposed resolution says
Replace "NORM" to "ageCoef * rateCoef". Remove references to NORM.
Add reference to ageCoef and rateCoef.
I believe this is incorrect. To denormalize an aged-rate calculation to bytes/second, it is divided by ageCoef*agingInterval.
Therefore replace NORM by 1/(ageCoef*agingInterval).
__________________________
MR0034: 11.6.5.6 - The state machine rows 10, 11, and 12 do not correspond to the row descriptions.
Row 12 does correspond to its description. Rows 10 and 11 don't; the discrepancy is whether WTR is enabled or not.
WTR is for delaying the clearing of an edge. In row 10 the edge is controlled by the neighbor so WRT should be disabled.
If the neighbor subsequently clears and the SD is still present, row 12 takes over.
In row 11 there is no edge so WTR should be disabled.
The descriptions for 10 and 11 are correct; in rows 10 and 11 wtr[ri].enabled should be changed to FALSE.
__________________________
MR0037: 11.6.5.6 - This allows an FS to be replaced by an MS
My interpretation is that it is assumed that the system admin knows what it wants and there is no need for the MAC state machine to check it.
For example, if adminReqProtection=FS is replaced by adminReqProtection=MS, it implies that the system has done it's qualification of the changed request.
All the MAC has to do is check whether the MS is feasible for the current ring state, as it always does.
The change FS -> MS is implicitly FS -> IDLE -> MS. If the MS fails then the result is as if the request were IDLE.
So I think Table 11.15 is OK.